What the milkman can teach us about inventory management excellence

What the milkman can teach us about inventory management excellence

Challenge

Whenever the topic of forecasting comes up in a group of supply chain professionals, eyes roll, and heads shake. I’m sure that you know what I’m talking about. As we all know, forecasts are notoriously inaccurate. In fact, that is the only consistent aspect of forecasts – they are always inaccurate.

However, you can’t just blame your commercial team for your supply chain woes. What about your bills of material, production plans, inventory on hand etc.? Is it at the required 95% accuracy level, or do you even know?

Unfortunately, most supply chain decisions rely upon all this inaccurate data, resulting in excess inventory or inventory shortages depending upon the direction of the inaccuracy. So, what are we to do about it?

Solution

Although there are several techniques that will make your forecasts and key planning data more accurate, kanban is a technique that, with the right discipline, minimizes the reliance of your planning processes on this data.

Kanban gives replenishment messages when inventory has actually reached the reorder point level, not when it is projected to reach it. In other words, it is a “pull” system, versus the traditional “push” system.

To illustrate, let’s look at an easy-to-understand example. For those of you that are old enough, you may recall the milkman who would deliver milk to your doorstep every day. Even though he didn’t realize it, he was also using the “pull” principle.

No alt text provided for this image

The process went something like this:

  1. You estimated how many bottles of milk (kanbans) will be consumed daily.
  2. Milkman delivers that many bottles (replenishment)
  3. Consumer drinks milk and leaves out the empty bottles (inventory usage)
  4. Milkman replaces the empty bottles with full ones (replenishment)
  5. Repeat

It worked like clockwork every day, but what if this were a “push” system, instead? Now the same number of bottles would be delivered every day, regardless of usage. And if the family goes on vacation? Yep, they would come home to a doorstep full of sour milk!

The big difference is the control on the amount of inventory in the system. In the example above, if you decided to have 4 bottles (kanban level), then your inventory can not exceed 4 bottles, whereas in a “push” system, there is no limit! I’ve seen organizations with a year’s worth of inventory, for an item with a 1 week lead time.

Another benefit is that since it is a visual system, input errors in your ERP system won’t screw things up with a premature replenishment signal (false positive) or give no signal when one is badly needed (false negative). In this way, kanban means less inventory and less supply risk.

Results

We transitioned a manufacturer of expensive bioanalytical instruments from MRP to kanban. These instruments had over 600 items in the bill of material, and if one was unavailable then production was shut down. With kanban the number of shortages reduced by 37% and inventory reduced by 30% within 12 months.

At a large medical device manufacturer, we implemented kanban for finished goods. The production site assembled these finished goods, which were stored refrigerated in a local warehouse, which was at full capacity. Management had committed to a $2MM larger refrigerator, but with nothing more than a whiteboard, sticky dots, and magnets, finished goods inventory reduced by over 30%, and the $2MM cost was avoided!

To learn more about BioSupply, please visit their website https://biosupplyconsulting.com/

No alt text provided for this image
No alt text provided for this image



Arvind Yadav

Principal - Digital Supply Chain (SAP IBP, S/4HANA) at DXC Technology

3 天前

Hi Steve, Thank you for sharing your valuable insights on Kanban and the Pull system. I appreciate the perspective you’ve provided. That said, I have a slightly different view on some of the points you’ve mentioned. As an example, you noted that a Push system lacks an upper limit on inventory. However, most ERP/MRP systems have long had a "Replenish Up to Max Level" parameter, which allows businesses to cap inventory at a defined level—similar to the four-bottle example you provided. This capability has been available since the early days of MRP, first manually and later through software since 1956. In practice, there is no single "best" approach that fits all scenarios. Modern supply chain planning offers a range of methodologies, from Classic MRP and MRP Live to DDMRP, optimization-based planning, demand sensing, and AI/ML-driven forecasting. The choice of approach depends on various factors, including an organization’s business goals, service levels, supply network, and the level of demand and supply variability. Looking forward to hearing your thoughts! Cheers Arvind

回复

要查看或添加评论,请登录

Steve Clarke的更多文章